diff mbox series

ACPI: LPSS: Add 80862289 ACPI HID for second PWM controller on Cherry Trail

Message ID 20230417211128.588008-1-hdegoede@redhat.com
State Superseded
Headers show
Series ACPI: LPSS: Add 80862289 ACPI HID for second PWM controller on Cherry Trail | expand

Commit Message

Hans de Goede April 17, 2023, 9:11 p.m. UTC
On some Cherry Trail devices the second PWM controller uses
80862289 as ACPI HID, rather then using 80862288 as is done
for both controllers on most models.

Add the missing 80862289 ACPI HID, note this uses its own
lpss_device_desc, without ".setup = bsw_pwm_setup" so that
the pwm_lookup is not added for it.
On devices where both controllers use the 80862288 HID bsw_pwm_setup()
does a UID check to avoid registering the lookup for the second
controller but that will not work here.

Adding the missing id fixes the second PWM controller no longer
working after the entire LPSS1 island has been in D3 at least
once, which causes the contents of the LPSS private registers
to get lost. Adding the HID makes acpi_lpss restore these
when the controller moves from D3 to D0.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
---
 drivers/acpi/acpi_lpss.c | 7 +++++++
 1 file changed, 7 insertions(+)

Comments

Mika Westerberg April 18, 2023, 6:07 a.m. UTC | #1
Hi Hans,

On Mon, Apr 17, 2023 at 11:11:28PM +0200, Hans de Goede wrote:
> On some Cherry Trail devices the second PWM controller uses
> 80862289 as ACPI HID, rather then using 80862288 as is done

ACPI _HID (ditto in $subjet)

> for both controllers on most models.
> 
> Add the missing 80862289 ACPI HID, note this uses its own

ACPI _HID

> lpss_device_desc, without ".setup = bsw_pwm_setup" so that
> the pwm_lookup is not added for it.
> On devices where both controllers use the 80862288 HID bsw_pwm_setup()

_HID

> does a UID check to avoid registering the lookup for the second

_UID

> controller but that will not work here.
> 
> Adding the missing id fixes the second PWM controller no longer
> working after the entire LPSS1 island has been in D3 at least
> once, which causes the contents of the LPSS private registers
> to get lost. Adding the HID makes acpi_lpss restore these

_HID

> when the controller moves from D3 to D0.
> 
> Signed-off-by: Hans de Goede <hdegoede@redhat.com>

With the above changes, feel free to add

Reviewed-by: Mika Westerberg <mika.westerberg@linux.intel.com>
diff mbox series

Patch

diff --git a/drivers/acpi/acpi_lpss.c b/drivers/acpi/acpi_lpss.c
index f08ffa75f4a7..77186f084d3a 100644
--- a/drivers/acpi/acpi_lpss.c
+++ b/drivers/acpi/acpi_lpss.c
@@ -271,6 +271,12 @@  static const struct lpss_device_desc bsw_pwm_dev_desc = {
 	.resume_from_noirq = true,
 };
 
+static const struct lpss_device_desc bsw_pwm2_dev_desc = {
+	.flags = LPSS_SAVE_CTX_ONCE | LPSS_NO_D3_DELAY,
+	.prv_offset = 0x800,
+	.resume_from_noirq = true,
+};
+
 static const struct lpss_device_desc byt_uart_dev_desc = {
 	.flags = LPSS_CLK | LPSS_CLK_GATE | LPSS_CLK_DIVIDER | LPSS_SAVE_CTX,
 	.clk_con_id = "baudclk",
@@ -368,6 +374,7 @@  static const struct acpi_device_id acpi_lpss_device_ids[] = {
 	/* Braswell LPSS devices */
 	{ "80862286", LPSS_ADDR(lpss_dma_desc) },
 	{ "80862288", LPSS_ADDR(bsw_pwm_dev_desc) },
+	{ "80862289", LPSS_ADDR(bsw_pwm2_dev_desc) },
 	{ "8086228A", LPSS_ADDR(bsw_uart_dev_desc) },
 	{ "8086228E", LPSS_ADDR(bsw_spi_dev_desc) },
 	{ "808622C0", LPSS_ADDR(lpss_dma_desc) },